© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 15342
    Prev
    Next
    INDEX
    Explanations

    code structure terminations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     and
    -1.14
     všechno
    -0.93
     slecht
    -0.92
    nému
    -0.92
    </u>
    -0.92
     australiano
    -0.90
    calça
    -0.90
    verez
    -0.88
     from
    -0.87
     Gnaden
    -0.85
    POSITIVE LOGITS
    iramente
    1.16
    ícias
    1.10
    zación
    1.02
    raeli
    1.02
     vinho
    1.00
     complexo
    0.99
     sapiens
    0.99
    FFECT
    0.98
    ezers
    0.97
    licante
    0.96
    Activations Density 0.008%

    No Known Activations